摘要
城市道路信息在智能交通中扮演着重要角色,其具有明显颜色特征,但两边树木和建筑物会遮挡道路,影响道路信息的精确提取。为了提取完整的道路信息,本文提出一种半自动道路信息提取算法。首先,人机交互方式选取道路点,根据颜色特征提取粗略道路信息,部分道路信息存在缺失;其次,采用分治法,将粗略道路信息图分成四个图像块,以提高计算的效率;然后,对图像块中不连续区域进行统计,并寻找每个区域到其他区域的最小距离,设计并实现区域合并算法,将不连续的道路合并为完整道路信息,为智慧城市服务。仿真效果显示该算法有效性。
Urban road information plays an important role in intelligent transportation, which has obvious color characteristics. While the road is blocked by trees and buildings on both sides, which will affect the accurate extraction of road information.This paper presents a semi-automatic road information extraction algorithm, which can extract accurate road information. Firstly,the road points are selected by human-computer interaction, and rough road information is extracted according to color features.Secondly, the divide and conquer method is used to divide rough road information map into four image blocks to improve calculation efficiency. Then, the discontinuous regions in the image block are counted, and the minimum distance between each region and other regions is found. The region merging algorithm is designed and implemented to merge the discontinuous roads into complete road information, which serves the smart city. Simulation results show that the algorithm is effective.
作者
肖驰
田小霞
Xiao Chi;Tian Xiaoxia(School of Computer and Information Engineering,Hanshan Normal University,Chaozhou 521041)
出处
《现代计算机》
2022年第5期98-102,共5页
Modern Computer
基金
韩山师范学院一般项目(LY201801)
潮州市科技局项目(2018GY20)
广东省教育厅项目(2017KTSC121)。
关键词
颜色特征
道路提取
分治法
区域合并
遥感图像
color feature
road extraction
divide and conquer algorithm
region merge
remote sensing image
